Web-based artificial intelligence to predict cognitive impairment following stroke: A multicenter study.

BACKGROUND AND PURPOSE: Post-stroke cognitive impairment (PSCI) is highly prevalent in modern society. However, there is limited study implying an accurate and explainable machine learning model to predict PSCI. The aim of this study is to develop and validate a web-based artificial intelligence (AI) tool for predicting PSCI. METHODS: The retrospective cohort study design was conducted to develop and validate a web-based prediction model. Adults who experienced a stroke between January 1, 2004, and September 30, 2017, were enrolled, and patients with PSCI were followed up from the stroke index date until their last follow-up. The model's performance metrics, including accuracy, area under the curve (AUC), recall, precision, and F1 score, were compared. RESULTS: A total of 3209 stroke patients were included in the study. The model demonstrated an accuracy of 0.8793, AUC of 0.9200, recall of 0.6332, precision of 0.9664, and F1 score of 0.7651. In the external validation phase, the accuracy improved to 0.9039, AUC to 0.9094, recall to 0.7457, precision to 0.9168, and F1 score to 0.8224. The final model can be accessed at https://psci-calculator.my.id/. CONCLUSION: Our results are able to produce a user-friendly interface that is useful for health practitioners to perform early prediction on PSCI. These findings also suggest that the provided AI model is reliable and can serve as a roadmap for future studies using AI models in a clinical setting.

Machine learning approaches for predicting 5-year breast cancer survival: A multicenter study.

The study used clinical data to develop a prediction model for breast cancer survival. Breast cancer prognostic factors were explored using machine learning techniques. We conducted a retrospective study using data from the Taipei Medical University Clinical Research Database, which contains electronic medical records from three affiliated hospitals in Taiwan. The study included female patients aged over 20 years who were diagnosed with primary breast cancer and had medical records in hospitals between January 1, 2009 and December 31, 2020. The data were divided into training and external testing datasets. Nine different machine learning algorithms were applied to develop the models. The performances of the algorithms were measured using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1-score. A total of 3914 patients were included in the study. The highest AUC of 0.95 was observed with the artificial neural network model (accuracy, 0.90; sensitivity, 0.71; specificity, 0.73; PPV, 0.28; NPV, 0.94; and F1-score, 0.37). Other models showed relatively high AUC, ranging from 0.75 to 0.83. According to the optimal model results, cancer stage, tumor size, diagnosis age, surgery, and body mass index were the most critical factors for predicting breast cancer survival. The study successfully established accurate 5-year survival predictive models for breast cancer. Furthermore, the study found key factors that could affect breast cancer survival in Taiwanese women. Its results might be used as a reference for the clinical practice of breast cancer treatment.

Machine Learning Approaches for Predicting Psoriatic Arthritis Risk Using Electronic Medical Records: Population-Based Study.

BACKGROUND: Psoriasis (PsO) is a chronic, systemic, immune-mediated disease with multiorgan involvement. Psoriatic arthritis (PsA) is an inflammatory arthritis that is present in 6%-42% of patients with PsO. Approximately 15% of patients with PsO have undiagnosed PsA. Predicting patients with a risk of PsA is crucial for providing them with early examination and treatment that can prevent irreversible disease progression and function loss. OBJECTIVE: The aim of this study was to develop and validate a prediction model for PsA based on chronological large-scale and multidimensional electronic medical records using a machine learning algorithm. METHODS: This case-control study used Taiwan's National Health Insurance Research Database from January 1, 1999, to December 31, 2013. The original data set was split into training and holdout data sets in an 80:20 ratio. A convolutional neural network was used to develop a prediction model. This model used 2.5-year diagnostic and medical records (inpatient and outpatient) with temporal-sequential information to predict the risk of PsA for a given patient within the next 6 months. The model was developed and cross-validated using the training data and was tested using the holdout data. An occlusion sensitivity analysis was performed to identify the important features of the model. RESULTS: The prediction model included a total of 443 patients with PsA with earlier diagnosis of PsO and 1772 patients with PsO without PsA for the control group. The 6-month PsA risk prediction model that uses sequential diagnostic and drug prescription information as a temporal phenomic map yielded an area under the receiver operating characteristic curve of 0.70 (95% CI 0.559-0.833), a mean sensitivity of 0.80 (SD 0.11), a mean specificity of 0.60 (SD 0.04), and a mean negative predictive value of 0.93 (SD 0.04). CONCLUSIONS: The findings of this study suggest that the risk prediction model can identify patients with PsO at a high risk of PsA. This model may help health care professionals to prioritize treatment for target high-risk populations and prevent irreversible disease progression and functional loss.

Renin-Angiotensin-Aldosterone System Inhibitors and Development of Gynecologic Cancers: A 23 Million Individual Population-Based Study.

The chronic receipt of renin-angiotensin-aldosterone system (RAAS) inhibitors including angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) have been assumed to be associated with a significant decrease in overall gynecologic cancer risks. This study aimed to investigate the associations of long-term RAAS inhibitors use with gynecologic cancer risks. A large population-based case-control study was conducted from claim databases of Taiwan's Health and Welfare Data Science Center (2000-2016) and linked with Taiwan Cancer Registry (1979-2016). Each eligible case was matched with four controls using propensity matching score method for age, sex, month, and year of diagnosis. We applied conditional logistic regression with 95% confidence intervals to identify the associations of RAAS inhibitors use with gynecologic cancer risks. The statistical significance threshold was p < 0.05. A total of 97,736 gynecologic cancer cases were identified and matched with 390,944 controls. The adjusted odds ratio for RAAS inhibitors use and overall gynecologic cancer was 0.87 (95% CI: 0.85-0.89). Cervical cancer risk was found to be significantly decreased in the groups aged 20-39 years (aOR: 0.70, 95% CI: 0.58-0.85), 40-64 years (aOR: 0.77, 95% CI: 0.74-0.81), ≥65 years (aOR: 0.87, 95% CI: 0.83-0.91), and overall (aOR: 0.81, 95% CI: 0.79-0.84). Ovarian cancer risk was significantly lower in the groups aged 40-64 years (aOR: 0.76, 95% CI: 0.69-0.82), ≥65 years (aOR: 0.83, 95% CI: 0.75-092), and overall (aOR: 0.79, 95% CI: 0.74-0.84). However, a significantly increased endometrial cancer risk was observed in users aged 20-39 years (aOR: 2.54, 95% CI: 1.79-3.61), 40-64 years (aOR: 1.08, 95% CI: 1.02-1.14), and overall (aOR: 1.06, 95% CI: 1.01-1.11). There were significantly reduced risks of gynecologic cancers with ACEIs users in the groups aged 40-64 years (aOR: 0.88, 95% CI: 0.84-0.91), ≥65 years (aOR: 0.87, 95% CI: 0.83-0.90), and overall (aOR: 0.88, 95% CI: 0.85-0.80), and ARBs users aged 40-64 years (aOR: 0.91, 95% CI: 0.86-0.95). Our case-control study demonstrated that RAAS inhibitors use was associated with a significant decrease in overall gynecologic cancer risks. RAAS inhibitors exposure had lower associations with cervical and ovarian cancer risks, and increased endometrial cancer risk. ACEIs/ARBs use was found to have a preventive effect against gynecologic cancers. Future clinical research is needed to establish causality.

Risk of cancer in long-term levothyroxine users: Retrospective population-based study.

Levothyroxine is a widely prescribed medication for the treatment of an underactive thyroid. The relationship between levothyroxine use and cancer risk is largely underdetermined. To investigate the magnitude of the possible association between levothyroxine use and cancer risk, this retrospective case-control study was conducted using Taiwan's Health and Welfare Data Science Center database. Cases were defined as all patients who were aged ≥20 years and had a first-time diagnosis for cancer at any site for the period between 2001 and 2011. Multivariable conditional logistic regression models were used to calculate an adjusted odds ratio (AOR) to reduce potential confounding factors. A total of 601 733 cases and 2 406 932 controls were included in the current study. Levothyroxine users showed a 50% higher risk of cancer at any site (AOR: 1.50, 95% CI: 1.46-1.54; P < .0001) compared with non-users. Significant increased risks were also observed for brain cancer (AOR: 1.90, 95% CI: 1.48-2.44; P < .0001), skin cancer (AOR: 1.42, 95% CI: 1.17-1.72; P < .0001), pancreatic cancer (AOR: 1.27, 95% CI: 1.01-1.60; P = .03), and female breast cancer (AOR: 1.24, 95% CI: 1.15-1.33; P < .0001). Our study results showed that levothyroxine use was significantly associated with an increased risk of cancer, particularly brain, skin, pancreatic, and female breast cancers. Levothyroxine remains a highly effective therapy for hypothyroidism; therefore, physicians should carefully consider levothyroxine therapy and monitor patients' condition to avoid negative outcomes. Additional studies are needed to confirm these findings and to evaluate the potential biological mechanisms.

Statins use and its impact in EGFR-TKIs resistance to prolong the survival of lung cancer patients: A Cancer registry cohort study in Taiwan.

Statins have been shown to be a beneficial treatment as chemotherapy and target therapy for lung cancer. This study aimed to investigate the effectiveness of statins in combination with epidermal growth factor receptor-tyrosine kinase inhibitor therapy for the resistance and mortality of lung cancer patients. A population-based cohort study was conducted using the Taiwan Cancer Registry database. From January 1, 2007, to December 31, 2012, in total 792 non-statins and 41 statins users who had undergone EGFR-TKIs treatment were included in this study. All patients were monitored until the event of death or when changed to another therapy. Kaplan-Meier estimators and Cox proportional hazards regression models were used to calculate overall survival. We found that the mortality was significantly lower in patients in the statins group compared with patients in the non-statins group (4-y cumulative mortality, 77.3%; 95% confidence interval (CI), 36.6%-81.4% vs. 85.5%; 95% CI, 78.5%-98%; P = .004). Statin use was associated with a reduced risk of death in patients the group who had tumor sizes <3 cm (hazard ratio [HR], 0.51, 95% CI, 0.29-0.89) and for patients in the group who had CCI scores <3 (HR, 0.6; 95% CI, 0.41-0.88; P = .009). In our study, statins were found to be associated with prolonged survival time in patients with lung cancer who were treated with EGFR-TKIs and played a synergistic anticancer role.

Meta-analysis of proton pump inhibitors induced risk of community-acquired pneumonia.

PURPOSE: Proton pump inhibitors (PPIs), one of the most widely used medications, are commonly used to suppress several acid-related upper gastrointestinal disorders. Acid-suppressing medication use could be associated with increased risk of community-acquired pneumonia (CAP), although the results of clinical studies have been conflicting. DATA SOURCES: A comprehensive search of MEDLINE, EMBASE and Cochrane library and Database of Systematic Reviews from the earliest available online year of indexing up to October 2018. STUDY SELECTION: We performed a systematic review and meta-analysis of observational studies to evaluate the risk of PPI use on CAP outcomes. DATA EXTRACTION: Included study location, design, population, the prevalence of CAP, comparison group and other confounders. We calculated pooled odds ratio (OR) using a random-effects meta-analysis. RESULTS OF DATA SYNTHESIS: Of the 2577 studies screening, 11 papers were included in the systematic review and 7 studies with 65 590 CAP cases were included in the random-effects meta-analysis. In current PPI users, pooled OR for CAP was 1.86 (95% confidence interval (CI), 1.30-2.66), and in the case of recent users, OR for CAP was 1.66 (95% CI, 1.22-2.25). In the subgroup analysis of CAP, significance association is also observed in both high-dose and low-dose PPI therapy. When stratified by duration of exposure, 3-6 months PPIs users group was associated with increased risk of developing CAP (OR, 2.05; 95% CI, 1.22-3.45). There was a statistically significant association between the PPI users and the rate of hospitalization (OR, 2.59; 95% CI, 1.83-3.66). CONCLUSION: We found possible evidence linking PPI use to an increased risk of CAP. More randomized controlled studies are warranted to clarify an understanding of the association between PPI use and risk of CAP because observational studies cannot clarify whether the observed epidemiologic association is a causal effect or a result of unmeasured/residual confounding.

Levothyroxine use and the risk of breast cancer: a nation-wide population-based case-control study.

PURPOSE: To investigate whether the use of levothyroxine was associated with breast cancer risk. METHODS: We conducted a population-based case-control study in Taiwan. Cases consisted of all patients who were aged 20 years and older, and had a first-time diagnosis of breast cancer for the period between 2001 and 2011. The controls were matched to the cases by age, sex, year, and month of diagnosis. Adjusted odd ratios (ORs) and 95% confidence intervals (CIs) were estimated by a conditional logistic regression. RESULTS: We examined 65,491 breast cancer cases and 261,964 controls. We found that use of levothyroxine was associated with a significant increase in breast cancer risk (OR 1.24, 95% CI 1.15-1.33; P < 0.001). Compared with no use levothyroxine, the adjusted odd ratio was 1.22 (95% CI 1.11-1.35; P = 0.01) for the group having been prescribed levothyroxine 2 months to 1 year, and 1.26 (95% CI 1.12-1.41; P < 0.01) for the group with more than 1 year. When stratified by age, the adjusted odd ratio was 1.45 (95% CI 1.23-1.71; P < 0.01) for the patients aged 65 years or more and 1.19 (95% CI 1.09-1.29, P < 0.01) for the patients aged less than 65 years. CONCLUSION: The results of the present study are the first to suggest that levothyroxine use increased the risk of breast cancer. However, a larger long-term prospective randomized-controlled trial specifically designed to assess the effect of levothyroxine use on the risk of developing breast cancer is needed.

Exploring the Association between Statin Use and the Risk of Parkinson's Disease: A Meta-Analysis of Observational Studies.

BACKGROUND: Parkinson's disease (PD) is a progressive disorder of the central nervous system. The prevalence of PD varies considerably by age group; it has a higher prevalence in patients aged 60 years and more. Several studies have shown that statin, a cholesterol-lowering medication, reduces the risk of developing PD, but evidence for this is so far inconclusive. The objective of this study is to evaluate the association between statin use and the risk of developing PD. METHODS: PubMed, EMBASE, and the bibliographies of articles were searched for studies published between January 1, 1990, and January 1, 2017, which reported on the association between statin use and PD. Articles were included if they (1) were published in English, (2) reported patients treated with statin, and the outcome of interest was PD, (3) provided OR/HR with 95% CI or sufficient information to calculate the 95% CI. All abstracts, full-text articles, and sources were reviewed, with duplicate data excluded. Summary relative risk (RRs) with 95% CI was pooled using a random-effects model. Subgroup and sensitivity analyses were also conducted. RESULTS: We selected 16 out of 529 unique abstracts for full-text review using our selection criteria, and 13 out of these 16 studies, comprising 4,877,059 persons, met all of our inclusion criteria. The overall pooled RR of PD was 0.70 (95% CI 0.58-0.84) with significant heterogeneity between estimates (I2 = 93.41%, p = 0.000) for the random-effects model. In subgroup analysis, the greater decreased risk was found in studies from Asia (RR 0.62 95% CI 0.51-0.76), whereas a moderate reduction was observed in studies from North America (RR 0.69 95% CI 0.47-1.00), but less reduction was observed in studies from Europe (RR 0.86 95% CI 0.80-0.92). Also, long-term statin use, simvastatin, and atorvastatin showed a higher rate of reduction with significance heterogeneity. CONCLUSION: Our results showed that statin use is significantly associated with a lower risk of developing PD. Physicians should consider statin drug therapy, monitor its outcomes, and empower their patients to improve their knowledge, therapeutic outcomes, and quality of life. However, preventive measures and their associated mechanisms must be further assessed and explored.

Benzodiazepines use and breast cancer risk: A population-based study and gene expression profiling evidence.

The aim of this study was to investigate whether long-term use of Benzodiazepines (BZDs) is associated with breast cancer risk through the combination of population-based observational and gene expression profiling evidence. We conducted a population-based case-control study by using 1998 to 2009year Taiwan National Health Insurance Research Database and investigated the association between BZDs use and breast cancer risk. We selected subjects age of >20years old and six eligible controls matched for age, sex and the index date (i.e., free of any cancer at the case diagnosis date) by using propensity scores. A bioinformatics analysis approach was also performed for the identification of oncogenesis effects of BZDs on breast cancer. We used breast cancer gene expression data from the Cancer Genome Atlas and perturbagen signatures of BZDs from the Library of Integrated Cellular Signatures database in order to identify the oncogenesis effects of BZDs on breast cancer. We found evidence of increased breast cancer risk for diazepam (OR, 1.16; 95%CI, 0.95-1.42; connectivity score [CS], 0.3016), zolpidem (OR, 1.11; 95%CI, 0.95-1.30; CS, 0.2738), but not for lorazepam (OR, 1.04; 95%CI, 0.89-1.23; CS, -0.2952) consistently in both methods. The finding for alparazolam was contradictory from the two methods. Diazepam and zolpidem trends showed association, although not statistically significant, with breast cancer risk in both epidemiological and bioinformatics analyses outcomes. The methodological value of our study is in introducing the way of combining epidemiological and bioinformatics approaches in order to answer a common scientific question. Combining the two approaches would be a substantial step towards uncovering, validation and further application of previously unknown scientific knowledge to the emerging field of precision medicine informatics.

Exploring association between statin use and breast cancer risk: an updated meta-analysis.

PURPOSE: The benefits of statin treatment for preventing cardiac disease are well established. However, preclinical studies suggested that statins may influence mammary cancer growth, but the clinical evidence is still inconsistent. We, therefore, performed an updated meta-analysis to provide a precise estimate of the risk of breast cancer in individuals undergoing statin therapy. METHODS: For this meta-analysis, we searched PubMed, the Cochrane Library, Web of Science, Embase, and CINAHL for published studies up to January 31, 2017. Articles were included if they (1) were published in English; (2) had an observational study design with individual-level exposure and outcome data, examined the effect of statin therapy, and reported the incidence of breast cancer; and (3) reported estimates of either the relative risk, odds ratios, or hazard ratios with 95% confidence intervals (CIs). We used random-effect models to pool the estimates. RESULTS: Of 2754 unique abstracts, 39 were selected for full-text review, and 36 studies reporting on 121,399 patients met all inclusion criteria. The overall pooled risks of breast cancer in patients using statins were 0.94 (95% CI 0.86-1.03) in random-effect models with significant heterogeneity between estimates (I 2 = 83.79%, p = 0.0001). However, we also stratified by region, the duration of statin therapy, methodological design, statin properties, and individual stain use. CONCLUSIONS: Our results suggest that there is no association between statin use and breast cancer risk. However, observational studies cannot clarify whether the observed epidemiologic association is a causal effect or the result of some unmeasured confounding variable. Therefore, more research is needed.

Gender-based personalized pharmacotherapy: a systematic review.

PURPOSE: In general, male and female are prescribed the same amount of dosage even if most of the cases female required less dosage than male. Physicians are often facing problem on appropriate drug dosing, efficient treatment, and drug safety for a female in general. To identify and synthesize evidence about the effectiveness of gender-based therapy; provide the information to patients, providers, and health system intervention to ensure safety treatment; and minimize adverse effects. METHODS: We performed a systematic review to evaluate the effect of gender difference on pharmacotherapy. Published articles from January 1990 to December 2015 were identified using specific term in MEDLINE (PubMed), EMBASE, and the Cochrane library according to search strategies that strengthen the reporting of observational and clinical studies. RESULTS: Twenty-six studies fulfilled the inclusion criteria for this systematic review, yielding a total of 6309 subjects. We observed that female generally has a lower the gastric emptying time, gastric PH, lean body mass, and higher plasma volume, BMI, body fat, as well as reduce hepatic clearance, difference in activity of Cytochrome P450 enzyme, and metabolize drugs at different rate compared with male. Other significant factors such as conjugation, protein binding, absorption, and the renal elimination could not be ignored. However, these differences can lead to adverse effects in female especially for the pregnant, post-menopausal, and elderly women. CONCLUSION: This systematic review provides an evidence for the effectiveness of dosage difference to ensure safety and efficient treatment. Future studies on the current topic are, therefore, recommended to reduce the adverse effect of therapy.

Benzodiazepine Use and Risk of Dementia in the Elderly Population: A Systematic Review and Meta-Analysis.

BACKGROUND: Benzodiazepines are a widely used medication in developed countries, particularly among elderly patients. However, benzodiazepines are known to affect memory and cognition and might thus enhance the risk of dementia. The objective of this review is to synthesize evidence from observational studies that evaluated the association between benzodiazepines use and dementia risk. SUMMARY: We performed a systematic review and meta-analysis of controlled observational studies to evaluate the risk of benzodiazepines use on dementia outcome. All control observational studies that compared dementia outcome in patients with benzodiazepine use with a control group were included. We calculated pooled ORs using a random-effects model. Ten studies (of 3,696 studies identified) were included in the systematic review, of which 8 studies were included in random-effects meta-analysis and sensitivity analyses. Odds of dementia were 78% higher in those who used benzodiazepines compared with those who did not use benzodiazepines (OR 1.78; 95% CI 1.33-2.38). In subgroup analysis, the higher association was still found in the studies from Asia (OR 2.40; 95% CI 1.66-3.47) whereas a moderate association was observed in the studies from North America and Europe (OR 1.49; 95% CI 1.34-1.65 and OR 1.43; 95% CI 1.16-1.75). Also, diabetics, hypertension, cardiac disease, and statin drugs were associated with increased risk of dementia but negative association was observed in the case of body mass index. There was significant statistical and clinical heterogeneity among studies for the main analysis and most of the sensitivity analyses. There was significant statistical and clinical heterogeneity among the studies for the main analysis and most of the sensitivity analyses. Key Messages: Our results suggest that benzodiazepine use is significantly associated with dementia risk. However, observational studies cannot clarify whether the observed epidemiologic association is a causal effect or the result of some unmeasured confounding variable. Therefore, more research is needed.

An automated technique to identify potential inappropriate traditional Chinese medicine (TCM) prescriptions.

PURPOSE: Medication errors such as potential inappropriate prescriptions would induce serious adverse drug events to patients. Information technology has the ability to prevent medication errors; however, the pharmacology of traditional Chinese medicine (TCM) is not as clear as in western medicine. The aim of this study was to apply the appropriateness of prescription (AOP) model to identify potential inappropriate TCM prescriptions. METHODS: We used the association rule of mining techniques to analyze 14.5 million prescriptions from the Taiwan National Health Insurance Research Database. The disease and TCM (DTCM) and traditional Chinese medicine-traditional Chinese medicine (TCMM) associations are computed by their co-occurrence, and the associations' strength was measured as Q-values, which often referred to as interestingness or life values. By considering the number of Q-values, the AOP model was applied to identify the inappropriate prescriptions. Afterwards, three traditional Chinese physicians evaluated 1920 prescriptions and validated the detected outcomes from the AOP model. RESULT: Out of 1920 prescriptions, 97.1% of positive predictive value and 19.5% of negative predictive value were shown by the system as compared with those by experts. The sensitivity analysis indicated that the negative predictive value could improve up to 27.5% when the model's threshold changed to 0.4. CONCLUSION: We successfully applied the AOP model to automatically identify potential inappropriate TCM prescriptions. This model could be a potential TCM clinical decision support system in order to improve drug safety and quality of care.

A richly interactive exploratory data analysis and visualization tool using electronic medical records.

BACKGROUND: Electronic medical records (EMRs) contain vast amounts of data that is of great interest to physicians, clinical researchers, and medial policy makers. As the size, complexity, and accessibility of EMRs grow, the ability to extract meaningful information from them has become an increasingly important problem to solve. METHODS: We develop a standardized data analysis process to support cohort study with a focus on a particular disease. We use an interactive divide-and-conquer approach to classify patients into relatively uniform within each group. It is a repetitive process enabling the user to divide the data into homogeneous subsets that can be visually examined, compared, and refined. The final visualization was driven by the transformed data, and user feedback direct to the corresponding operators which completed the repetitive process. The output results are shown in a Sankey diagram-style timeline, which is a particular kind of flow diagram for showing factors' states and transitions over time. RESULTS: This paper presented a visually rich, interactive web-based application, which could enable researchers to study any cohorts over time by using EMR data. The resulting visualizations help uncover hidden information in the data, compare differences between patient groups, determine critical factors that influence a particular disease, and help direct further analyses. We introduced and demonstrated this tool by using EMRs of 14,567 Chronic Kidney Disease (CKD) patients. CONCLUSIONS: We developed a visual mining system to support exploratory data analysis of multi-dimensional categorical EMR data. By using CKD as a model of disease, it was assembled by automated correlational analysis and human-curated visual evaluation. The visualization methods such as Sankey diagram can reveal useful knowledge about the particular disease cohort and the trajectories of the disease over time.

Utilizing Health Information Technology to Support Universal Healthcare Delivery: Experience of a National Healthcare System.

Recent discussions have focused on using health information technology (HIT) to support goals related to universal healthcare delivery. These discussions have generally not reflected on the experience of countries with a large amount of experience using HIT to support universal healthcare on a national level. HIT was compared globally by using data from the Ministry of the Interior, Republic of China (Taiwan). Taiwan has been providing universal healthcare since 1995 and began to strategically implement HIT on a national level at that time. Today the national-level HIT system is more extensive in Taiwan than in many other countries and is used to aid administration, clinical care, and public health. The experience of Taiwan thus can provide an illustration of how HIT can be used to support universal healthcare delivery. In this article we present an overview of some key historical developments and successes in the adoption of HIT in Taiwan over a 17-year period, as well as some more recent developments. We use this experience to offer some strategic perspectives on how it can aid in the adoption of large-scale HIT systems and on how HIT can be used to support universal healthcare delivery.

Recent progress in machine learning approaches for predicting carcinogenicity in drug development.

INTRODUCTION: This review explores the transformative impact of machine learning (ML) on carcinogenicity prediction within drug development. It discusses the historical context and recent advancements, emphasizing the significance of ML methodologies in overcoming challenges related to data interpretation, ethical considerations, and regulatory acceptance. AREAS COVERED: The review comprehensively examines the integration of ML, deep learning, and diverse artificial intelligence (AI) approaches in various aspects of drug development safety assessments. It explores applications ranging from early-phase compound screening to clinical trial optimization, highlighting the versatility of ML in enhancing predictive accuracy and efficiency. EXPERT OPINION: Through the analysis of traditional approaches such as in vivo rodent bioassays and in vitro assays, the review underscores the limitations and resource intensity associated with these methods. It provides expert insights into how ML offers innovative solutions to address these challenges, revolutionizing safety assessments in drug development.

Machine learning-based prediction of medication refill adherence among first-time insulin users with type 2 diabetes.

AIMS: The prevalence of Type 2 Diabetes Mellitus (T2DM) is projected to be 7 % in 2030. Despite its need for long-term diabetes care, the adherence rate of injectable medications such as insulin is around 60 %, lower than the acceptable threshold of 80 %. This study aims to create classification models to predict insulin adherence among adult T2DM naïve insulin users. METHODS: Clinical data were extracted from Taipei Medical University Clinical Research Database (TMUCRD) from January 1st, 2004 to December 30th, 2020. A patient was regarded as adherent if his/her medication possession ratio (MPR) was at least 80 %. Seven domains of predictors were created, including demographics, baseline medications, baseline comorbidities, baseline laboratory data, healthcare resource utilization, index insulins, and the concomitant non-insulin T2DM medications. We built two Xgboost models for internal and external testing respectively. RESULTS: Using a cohort of 4134 patients from Taiwan, our model achieved the Area Under the curve of the Receiver Operating Characteristic (AUROC) of the internal test was 0.782 and the AUROC of the external test was 0.771. the SHAP (SHapley Additive exPlanations) value showed that the number of prescribed medications, the number of outpatient visits, and laboratory data were predictive of future insulin adherence. CONCLUSIONS: This is the first study to predict adherence among adult naïve insulin users. The developed model is a potential clinical decision support tool to identify possible non-adherent patients for healthcare providers to design individualized education plans.

Machine-Learning Based Risk Assessment for Cancer Therapy-Related Cardiac Adverse Events Among Breast Cancer Patients.

The study aims to develop machine-learning models to predict cardiac adverse events in female breast cancer patients who receive adjuvant therapy. We selected breast cancer patients from a retrospective dataset of the Taipei Medical University Clinical Research Database and Taiwan Cancer Registry between January 2004 and December 2020. Patients were monitored at the date of prescribed chemo- and/or -target therapies until cardiac adverse events occurred during a year. Variables were used, including demographics, comorbidities, medications, and lab values. Logistics regression (LR) and artificial neural network (ANN) were used. The performance of the algorithms was measured by the area under the receiver operating characteristic curve (AUC). In total, 1321 patients (an equal 15039 visits) were included. The best performance of the artificial neural network (ANN) model was achieved with the AUC, precision, recall, and F1-score of 0.89, 0.14, 0.82, and 0.2, respectively. The most important features were a pre-existing cardiac disease, tumor size, estrogen receptor (ER), human epidermal growth factor receptor 2 (HER2), cancer stage, and age at index date. Further research is necessary to determine the feasibility of applying the algorithm in the clinical setting and explore whether this tool could improve care and outcomes.

Taipei Medical University Clinical Research Database: a collaborative hospital EHR database aligned with international common data standards.

OBJECTIVE: The objective of this paper is to provide a comprehensive overview of the development and features of the Taipei Medical University Clinical Research Database (TMUCRD), a repository of real-world data (RWD) derived from electronic health records (EHRs) and other sources. METHODS: TMUCRD was developed by integrating EHRs from three affiliated hospitals, including Taipei Medical University Hospital, Wan-Fang Hospital and Shuang-Ho Hospital. The data cover over 15 years and include diverse patient care information. The database was converted to the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM) for standardisation. RESULTS: TMUCRD comprises 89 tables (eg, 29 tables for each hospital and 2 linked tables), including demographics, diagnoses, medications, procedures and measurements, among others. It encompasses data from more than 4.15 million patients with various medical records, spanning from the year 2004 to 2021. The dataset offers insights into disease prevalence, medication usage, laboratory tests and patient characteristics. DISCUSSION: TMUCRD stands out due to its unique advantages, including diverse data types, comprehensive patient information, linked mortality and cancer registry data, regular updates and a swift application process. Its compatibility with the OMOP CDM enhances its usability and interoperability. CONCLUSION: TMUCRD serves as a valuable resource for researchers and scholars interested in leveraging RWD for clinical research. Its availability and integration of diverse healthcare data contribute to a collaborative and data-driven approach to advancing medical knowledge and practice.